Could the Sauer engine top Second Life?

by SolidAndShade on 11/27/2006 06:28, 11 messages, last message: 11/28/2006 22:36, 2434 views, last view: 04/27/2024 13:01

Hey everyone,

I recently downloaded Second Life and explored the world for a bit, and was disappointed at the world\'s long loading times and restrictions on detailed objects. It seems to me that the Sauerbraten engine would be ideal for the creation of an online 3D chat environment with dynamic editing. Cube-based graphics would allow for short load times and complex structures, and the speed and fluidity of Sauerbraten beats the often clunky feel of Second Life. If you created a program with the Cube engine and a chat system, character model editing interface and advanced scripting system added on, I think it would be a formidable competitor for SL, especially if it were open source and anyone could put up a server. Thoughts.

   Board Index   

#1: Re: Easily.

by RealNitro on 11/27/2006 20:21, refers to #1

"Who knows..."

Is that a hint?

reply to this message

#2: Re: Easily.

by eihrul on 11/27/2006 20:26, refers to #1

No.

reply to this message

#3: ...

by Eivets on 11/27/2006 21:12

Hi,

I am working on a multiplayer role playing mod based on sauerbraten. It actually has some of the features you mention: Chat system, server side scripting and dynamic persistent world editing. That means, that you have the cooperative editing of sauerbraten and all the changes are stored persistently on the server and will survive client/server restarts. Changes are streamed to the client when connecting to a world. The user can be restricted, so that he can only edit his own parts of the world. In contrast to Second Life, the game supports not one huge map but unlimited (smaller) maps, between which the user can travel.

The biggest pitfall is the lighting: Once a user has edited a part of the world, the lightmaps are set to fullbright. That's by the way also the biggest problem of SL: They do not have lightmaps at all. Which is the reason, why the graphics look so boring.
I'm currently evaluating what can be done about that. Alternatives that come to mind would be to implement hardware shadows or to implement background calculation of lightmaps. Something like the patch mode but going on all the time in the background. I have a prototype for this which looks promising but crashes in some cases.

Apart from that, SL really full clumsy when moving. That's much better in sauerbraten.

If you want to have a look: w w w . r a d i g . c o m / s a u e r b r a t e n

reply to this message

#4: Re: Easily.

by Eivets on 11/27/2006 21:18, refers to #1

I agree partly. It's usually one of the two alternatives: You are absolutely bored because everything is basically the same, always. Or you are totally addicted and spend most of your time in the game. For professional reasons, I played several of those mmorpg and I got bored quickly in all cases. Maybe because I have a First Life (tm). But developing such a system is where the real fun is ;-)

reply to this message

#5: Re: ...

by SolidAndShade on 11/28/2006 02:00, refers to #3

Sounds interesting. I was thinking more about a 3D chatroom that the user can customize (basically what SL is) rather than an MMORPG. SL-type games don't involve grinding for money or levels, and combat only occurs in certain areas where scripts facilitate it, so the only addiction potential comes from the social interaction. The main thing that interests me about such games is creating new content, and the Cube engine's better suited to that in many ways than the SL engine. Second Life has a very low ceiling to the number of primitives you can have in a given area, whereas with Cube's landscape modeling system even highly complex forms can be quickly downloaded from a server.

reply to this message

#6: ..

by sinsky on 11/28/2006 11:11

I'm an addicted mmoer, and have thought a lot about it. There's a period in all weak men's life, when they must choose between drugs, alchohol, crime and murder, and mmos. If the man has a computer and internet connection, mmos are the natural choice. I've thought about the alternatives and still am not convinced that mmoing is best for me. But I can't move on to another branch, too weak now :)

Don't know how interesting developing a MMO can be, only heard that it's hell of a lot of work. About chatting in mmos, there's one difference compared to general chatting, in regard to people you don't know. While in "general chat" there's no way you can tell if what the person says is true, in mmo chat there's the alternative of talking about "game stuff" as opposed to "real life stuff". It's funny but "game talk" can get a lot more complicated (and interesting) than "real life talk", especially when it involves cheating and other activities that affect directly the players' survival in the game world. Normally cheating is illegal but on some UO freeshards for example it's allowed to some extent (you can make scripts to automate tasks while you're away from the PC for example), and I've found this to be the perfect environment for game talk. Of course, to outside people it would still seem that people involved in such talk are high, not making sense, etc., and "game survival" is not important compared to "real life survival". But it can be more interesting. It's a fact. That's why so many people choose to hang in game worlds when they can afford it.

reply to this message

#7: Re: ...

by Drakas on 11/28/2006 17:10, refers to #3

ok, what about mac and linux users?:P
Your program does not work here on Debian, just crashes using Wine!

reply to this message

#8: Re: ...

by Eivets on 11/28/2006 21:51, refers to #7

Hmm, I don't have a Mac, so I can't build for it. For linux, I did a build, but there, the collision detection does not work correctly. This has nothing to do with my part, because the original sauerbraten sources had the same problem, so it seems to be a build issue. Maybe wrong compiler (gcc 4) or wrong library versions. I will try again, later.

I would not expect to have success running a program like sauerbraten on wine. Does the original work???

Before you ask: at some point, I will make it open source, but not now. Needs some cleanup before...

reply to this message

#9: Re: ...

by Eivets on 11/28/2006 21:53, refers to #5

"Sounds interesting. I was thinking more about a 3D chatroom that the user can customize (basically what SL is) rather than an MMORPG"

Yeah, I'm currently evaluating, if it makes sense to develop more in that direction.. Fore a classical MMORPG, the cooperative editing does not make too much sense...

reply to this message

#10: Re: ...

by Eivets on 11/28/2006 21:53, refers to #5

"Sounds interesting. I was thinking more about a 3D chatroom that the user can customize (basically what SL is) rather than an MMORPG"

Yeah, I'm currently evaluating, if it makes sense to develop more in that direction.. Fore a classical MMORPG, the cooperative editing does not make too much sense...

reply to this message

#11: Hmm

by Pxtl on 11/28/2006 22:36

Well, keep in mind that one of the features of Second Life is its scriptability - for example, there are "conventional" MMORPGs within Second Life.

I suppose you could make a P2P-ish version of SL where players serve their own "zones" and very limited content is allowed to travel between zones. The problem SL faced was that the script system can be exploited to create plagues of monsters that hog system resources.

Probably the best approach would be something not like full-out SL, but simply a system more like conventional Sauerbraten, except that the map on one server is physically connected to a map on another server, so you can walk from one into another. "seeing" into the next server would just then be graphical, and only a minimal subset of scripting (primarily data-storage) would be allowed to function on servers that are not specifically running that script.

Either way, MMO-architecture is a freakishly hard task, and most intermediate approaches don't scale.

reply to this message

   Board Index